SwiftAudio performs caption-only distillation of a one-step TTA diffusion model by adapting VSD to audio with temporal smoothness regularization, achieving SOTA among one-step methods on AudioCaps and Clotho using ~45K captions.
Nonlinear total variation based noise removal algorithms
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
representative citing papers
Introduces nonconvex dual-TV regularizers based on transformed L1 for exponential-family tensor completion and proves error bounds of order O(n3 rt (max sk^2) log / n) that approach minimax rates up to O(max sk^2 / max(n1,n2)) gap.
The implicit regularizer of a Gaussian MMSE denoiser is an upper Moreau envelope of the negative log marginal density, making it 1-weakly convex and yielding O(1/√k) stationarity for plug-and-play gradient descent.
Reinforcement learning optimizes adaptive angle selection and dose allocation in sparse-view CT reconstruction, yielding better quality and defect detectability than uniform strategies under limited projections or dose.
citing papers explorer
-
SwiftAudio: Data-Efficient Caption-Only Distillation for One-Step Text-to-Audio Diffusion-based Generation
SwiftAudio performs caption-only distillation of a one-step TTA diffusion model by adapting VSD to audio with temporal smoothness regularization, achieving SOTA among one-step methods on AudioCaps and Clotho using ~45K captions.
-
Exponential-Family Tensor Completion via Nonconvex Dual Total-Variation Regularization
Introduces nonconvex dual-TV regularizers based on transformed L1 for exponential-family tensor completion and proves error bounds of order O(n3 rt (max sk^2) log / n) that approach minimax rates up to O(max sk^2 / max(n1,n2)) gap.
-
Nonasymptotic Convergence Rates for Plug-and-Play Methods With MMSE Denoisers
The implicit regularizer of a Gaussian MMSE denoiser is an upper Moreau envelope of the negative log marginal density, making it 1-weakly convex and yielding O(1/√k) stationarity for plug-and-play gradient descent.
-
Deep Reinforcement Learning for Optimizing Angle Selection and Dose Allocation in CT Reconstruction
Reinforcement learning optimizes adaptive angle selection and dose allocation in sparse-view CT reconstruction, yielding better quality and defect detectability than uniform strategies under limited projections or dose.